Publish Datasets and plots

Publish from the Study menu when a valid Job produced the result. Check the ID shown in the state line of the adopted study, especially if you typed a new ID after the last analysis.

Publish results

Choose Publish results to write a set of three Datasets. The Published canonical study Dataset message shows the profile ID. The other two have the same prefix and separate suffixes. The command is available when the viewer retains the adopted result and its valid Job. Upon reopening a Workspace with an unpublished result, Analyze could be re-executed.

IDs are replicate.<ID>.profile, replicate.<ID>.traces, replicate.<ID>.summary. Search for Datasets in the Session, in the module/curve.replicate_analysis/studies/<ID> branch. The publication creates a new set and refuses ID already present. If you want a separate set, choose a new Unique analysis ID, run Analyze and then publish. The publication is completed as a single operation.

profile

It has a line for each coordinate of the common grid. parameter grows from start to end of the domain. x_mean and y_mean form the consensus. For each axis you will find median, sample SD, conditional SEM, in/between/total components, counts, degrees of freedom and indicators of availability or truncation. The lower and higher limits are the average less or more SD, SEM or the total component, according to the name of the field.

Profile fieldsHow to read them
x_mean, y_mean, x_median, y_medianCentral values for each coordinate. Use the means to reconstruct the Consensus series.
x_std, y_std, x_sem, y_semSD sample and, if available, conditional SEM. They read in the axis unit.
x_within, x_between, x_totalSD within source, between sources and total. Consult the indicators before interpreting them.
x_n, x_n_sources, x_within_df, x_component_coefficientNumber of tracks, number of groups, degrees of freedom within the source and coefficient of the model with also different numerousities. For the number of independent observations stated use summary.
x_components_available, x_between_clamped and Y correspondingAvailability of components and negative estimation indicator among groups reported as zero. A missing value is distinct from 0 or 1.
lower_std, upper_std, lower_sem, upper_sem, lower_total, upper_total with x_ prefix or y_Descriptive limits per axis. For example y_lower_std is y_mean less y_std. Confidence and prediction intervals are unavailable.

covariance_xy reports the covariance between the two coordinates for that sample. Statistics and units are explained in Study designs and interpretation. A NaN in the columns of statistics indicates unavailability in that case.

traces

The rows are arranged for trace_index, starting at zero. In each track, the original stage = 0 rows appear first, then the stage = 1 rows, sampled on the grid. point_index starts at zero in each stage.

For stage 0, source_row is the canonical row ordinal in the source Dataset and parameter is missing. X and Y use the result units. For stage 1, source_row = -1 indicates that the interpolated sample has no original ordinal. parameter is the common coordinate. Use the provenance metadata to identify the Dataset and observation associated with a trace_index.

summary and metadata

The summary contains feature_index to identify the feature, trace_count for the number of tracks, independent_observation_count for the number of declared independent observations, common_parameter_start and common_parameter_end for the ends of the common parameter. In sensitivity the number of independent observations is 1. The statistical interpretations attached to traces and summary are references to the first coordinate of the profile. For the complete statistical point by point use profile.

In the Dataset browser, select the output. Metadata… allows you to control recipe, qualification, variable roles, origin rows and connected output IDs. Open active Dataset opens the standard table. You can reopen the outputs from the Session even after closing the viewer.

Publish Graph

Publish Graph creates an editable Plot v4 variant for the consensus X and Y values. If the profile is not yet published, the command also publishes the three Dataset first. The initial variant uses x_mean and y_mean. To find it after closing or changing Session, press Save Config in the main window of GX-TXT, or Session → Save configuration from the session tab. The variant was verified after this save and after the reopening of the Session. The visibility boxes of the canvas in Replicate Curves do not become series of the new variant. Read variant_id in the confirmation. In the main window open Analysis → Scientific Dataset Plot. Beside Scientific dataset 2D, press Variants…. Select that ID and press Edit to open the variant or Render to generate the chart.

In the tested version, saving the Workspace may fail after adopting a result or after Publish Graph. Published Datasets and plots remain available. To keep the recipe, follow the tested procedure for saving the recipe.